Automated surface models from light detection and ranging data for low-altitude microscale weather simulations

Enabling safe and performant drone operations over complex land surfaces hinges on accurate predictions of low-altitude wind variations. The dynamics of these winds are profoundly influenced by micro-features such as buildings, vegetation, and their associated material properties. Consequently, a meticulous representation of surface morphology at the meter scale becomes imperative for accurate forecasts. Over recent years, airborne light detection and ranging (LiDAR) has emerged as a technology that can produce detailed information about the land–atmosphere interface, providing coverage of entire countries at a sub-meter scale with centimeter-scale accuracy. This paves the way for high-precision surface models tailored for low-altitude microscale weather simulations, avoiding project-specific geospatial data acquisitions and labor-intensive processing. In this work, we present an automated surface model derived from the public 3D Elevation Program (3DEP) LiDAR dataset, which provides coverage for most of the continental United States. We also discuss how to accurately derive surface roughness, leaf area density, and other properties impacting microscale weather simulations from such a dataset and characterize associated uncertainties.